Curr. Sci. 32 245-247 (1963)
The trichromatic hypothesis
SIR C V RAMAN
%
"My design in this book is not to explain the properties of light by hypotheses, but
to propose and prove them by reason and experiments." So reads the opening
sentence of Newton's classical work on optics in the first-book of which the
foundations of the theory of colour were firmly laid and a bridge thus built
between the physics and the physiology of vision. Precisely what Newton had in
mind when he made the reference to hypotheses contained in this sentence may be
inferred from the following passage which appears towards the end of his book.
"As in mathematics so in natural philosophy, the investigation of difficult things
by the method of analysis ought ever to precede the method of composition. This
analysis consists in making experiments and observations, and in drawing
general conclusions from them by induction and admitting of no objection
against the conclusions but such as are taken from experiments, or other certain
truths. For hypotheses are not to be regarded in experimental philosophy." From
these remarks it is clear that the hypotheses which Newton had in mind are
assumptions of an arbitrary character not based on well established facts of
observation.
Newton's aversion to hypotheses was not unjustified. For, there is a danger in
adopting hypotheses when no real knowledge is available of the facts of the
subject. A species of self-deception then becomes possible leading one to beliefs
which are either wholly unjustified or else are only half-truths. Further, they are
liable to make one blind to facts which come to light later and which are
themselves a patent contradiction of the hypothetical assumptions. These
remarks are made here with special reference to the hypothesis originally put
forward by Thomas Young and now known and referred to generally as the
trichromatic theory of vision. That the theory is based on ad hoc assumptions and
not on any well-established facts will be made clear later on. It will suffice here to
mention that Young himself thought that the three primary sensations were those
of red, yellow and blue, and later changed over to red, green and violet as a better
choice. But before commenting any further on Young's hypothesis and its
subsequent history, it appears desirable in the first instance to state the actual
facts of the subject.
We may usefully begin by quoting in Newton's own words the conclusions
which he arrived at as the result of his studies on colour. In characteristic fashion,
he summed them up in two "definitions" which are reproduced below verbatim:
*
w
112'
C V RAMAN: F L O R A L * C O L O U R S A M P V I S U A L P E R C B P f I O N
Definition VII: The Light whose R a p are all alike Refrangible, f call Simple,
Homogeneal and ~imijar;/andthat whose Rays are some more Rejiangible than
others. I call Compound, Heterogeneal and Dissimilar. The former Light I call
Homogeneal, not because I would affirm it so in all respects, but because the. Rays
which agree in%efrangibility, agree at least in all those their other Properties
which I consider in the following Discourse.
Defmition VIII: The Colours of Homogeneal Lights, I cdll Primary, Homo geneal
and Simple; and those of Heterogeneal Lights, Heterogeheal and Compound. For
these are always compounded of the colours of Homogeneal Lights; as will
appear in the following Discourse.
Newton's ideas are very clearly expressed in the foregoing extracts. In the first
place, he recognised that the physically simplest forms of light-which we would
describe today as radiatiohs manifesting themselves as single sharp lines in the
spectrum-are also the exciters of the primary or simple physiological sensations
which aredhe pure colours of the spectrum, Newton also recognised that the
sensations excited by polychromatic light are compounded of these primary
sensations and are, therefore, necessarily of a more complex character.
That the sensations excited by monochromatic light are the primary physiological sensations and that these are quite as numerous as the colours which can
be perceived as distinct from each other in a pure spectrum is established by
varibus facts of observation. On no other basis can a reasonable explanation be
offered for the fact that our visual faculties enable us to distinguish between the
colour of closely adjacent regions in the spectrum. Indeed, in some parts of the
spectrum, a difference of as little as 10A in the wavelength of the light sufftces to
produce an observable difference in colour. Then again, if monochromatic light
be admixed with white light, we can still perceive and recognise the colour in such
admixture and what is perhaps even more significant is that our ability to
discriminate between the colours of closely adjacent regions of the spectrum is
not altered appreciably even when they are both admixed with substantial
proportions of white light.
Light, according to Newton's ideas expounded in the third book of his treatise,
is of a corpuscular nature. In other words, itgonsists of small bodies emitted by
the source of light, their sizes being different foi. the differently coloured rays of the
spectrum and altering continuously as we pass from one end of the visible
spectrum to the other. On this basis, the existence of a definite relationship
between the refrangibility of light and its observed colour is only to be expected.
To quote Newton's own words, "nothing more is requisite for producing all the
variety of colours, and degrees of refrangibility, than that the Rays of Light be
Bodies of different Sizes". It was inevitable, therefare, that Newton should
recognize the coldurs of the spectrum as the primary, hom'ogeneous, and simple
colours and the ~oloursof lights of different sorts mixed with each other as
heterogeneal and compound.
*'%
,
J
4,
TRICHROMATIC HYPOT~&SIS
113
The corpuscular view of the nature of light favoured by Newton fell into disrepute during the nineteenth century. It was no accident that the physicists who
were associated with the development of the wave-theory of light thought fit to
reject Newton's conclusions regarding colour and its perception and attempted
to replace them by dther assumptions of a hypothetical nature. They could not
have foreseen that all such attmpts were foredoomed to failure and that the
corpuscular concept of light would emerge once again, triumphantly viindicated.
The different sizes of the particles of light contemplated by Newton are replaced
by the different magnitudes of the energy-quanta which they represent.
It is a fact of observation that the eye can discern some 150 or more different
hues in the spectrum. Rejecting this as an inexplicable achievement of gur faculty
of vision, Young postulated that there are only three different "principal" colours
and that the rest are only derivatives. The question then arose, which three
colours should be chosen as the "principals". Young's first choice was that of the
colours red, yellow and blue. Later, he discarded these and adopted red, green
and violet as his favourites, He then drew an equilateral triangle having these
three colours at the vertices, white at the centre and the other spectral colours as '
points lying on its two sides.
Young's triangle of colours was just pure phantasy. For, later studies have
shown in the most conclusive manner that the pure colours of the spectrum swnd
in a category by themselves and that they cannot be equated to the result of any
superposition of other colours. This fact alone is sufficient to prove the
correctness of Newton's analysis of the subject of colour and is a shattering blow
to the ideas underlying the trichromatic hypothesis. But, as has been remarked
earlier, believers in ad hoc hypotheses do not readily admit defeat when
confronted by the discovery of new facts. They assiduously seek to find ways of
escape from the consequences of such discoveries.
One of the several ways in which it has been sought to bolster up a belief in the
validity of the trichromatic theory, instead of allowing it to join the limbo of
discarded hypotheses, has been to suggest that all observable colours could be
represented as equivalenl to the result of superposing three suitably chosen
colours in suitably chosen proportions. The equivalence is represented in the
form of an algebraic dquation, qmntities being introduced therein known
respectively as trichromatic coefficients and tristimulus values, suggestive of a
mysterious power and significance for the number three in colour theory.
Geometric representations have also been devised in which colours were
represented as points in a system of trilinear co-ordinates. A critical examination
of these representations of colour shows, however, that they are devoid of any real
physical significance. This becomes evident when it is remarked that in the XYZ
system which is gen<rallyadopted for the geometric representation of colour, the
vertices X, Y, Z afyfhe'triangle do not represent any real physical colours, the
entire triangle lying outside the area in which the points representing actual
colours lie. Indeed, these representations rJSd@nlittle more than that any actually
114
,
cv
R A M A N : FLORAL COLOURS A N D VISUAL PERCEPTION
observed colour resembles the result of superposing an achromatic sensation
upon a recognisable colour with a saturated hue, a fact which was known to and
stated quite clearly by Newton in his treatise.
The subject of the sensations excited on our visual organs by polychromatic
radiations is one of considerable interest and importance. One has only to recall
the vast number of possibilities included in the words "polychromatic radiation"
to appreciate that only observational data obtained on the widest possible basis
and by methods not influenced by bias of any sort could be expected to reveal the
real facts of the subject. So far from the trichromatic theory of vision having been
of any real assistance towards the understanding of this difficult and complex
field, it has only served to introduce error and confusion and stood in the way of
any real advances in knowledge.